Critical infrastructure: a social worlds study of values, design and resistance in Tor and the Tor community

Abstract

While cybercrime research has become well-established within criminology, there is very little criminological research which treats the infrastructures and platforms of the Internet as subjects of criminological enquiry. These are increasingly taking on responsibility for the governance of large populations of users, and the engineers and developers of these platforms are increasingly having to navigate problems of crime, harm, and policing. This thesis explores, through qualitative empirical research, an Internet infrastructure which has particularly faced these issues: the Tor Project, an anonymity network which gives millions of users around the world extremely strong protections against online surveillance and censorship. This has been an important tool for whistleblowers, journalists, and activists, however it has also become associated with a range of criminal uses, especially the rise of ‘cryptomarkets’, marketplaces for illegal services and goods accessible through the Tor network which are very difficult for law enforcement to shut down. I explore how the Tor community attempt to navigate these issues and how they make sense of the role Tor plays in society, drawing on interviews with members of the Tor community, including designers and developers, the people who maintain Tor’s infrastructure, and others in the Tor community, as well as extensive archival research in Tor’s online mailing list archives. I use frameworks from Science and Technology Studies, in particular, social worlds theory, to explore the values of the Tor community, how they attempt to materialise them through infrastructure, and the challenges they face in practice. The Tor community, rather than sharing a strong set of shared values, is in fact a dense thicket of contradictory values and meanings. Using social worlds theory, I distil this into three internally-coherent social worlds, each of which makes sense of the work Tor does differently, rooted in differing practices, sensibilities and understandings of the political salience of privacy technology. These are: the engineer social world, which views privacy as a structure, understanding privacy technologies as reshaping the topologies of power in information systems; the activist social world, which views privacy as a struggle and privacy technologies as part of a political movement; and finally, the infrastructuralist social world, which views privacy as a service and privacy technologies as the neutral facilitators of their users’ action. I explore the relationships between these three social worlds, how they have come into contact and conflict with one another, and how they have changed over the years. These each shape Tor’s material form, its attempts to cultivate resilience against disruption by powerful actors, and how it navigates its implications in crime, harm, and power in different ways, each of which I explore in detail. Although Tor represents an attempt to act in the domain of infrastructural power, it has found that doing politics through design and engineering relies on a lot of hidden work and complex negotiation in practice, spilling out into the domains of politics, administration, and governance and becoming caught up in the very technologies of control which it tries to subvert. I end the thesis with a discussion, drawing from my empirical research, of Tor’s place in the wider landscape of geopolitics and online power and how it makes sense of this. I argue that the challenges Tor faces are reflective of deeper tensions between freedom and control at the heart of liberal societies and how they are governed.